ABSTRACT
Sampling is defined as the selection of part of an aggregate of material to represent the whole aggregate. All rigorous sampling requires the subdivision of the material to be sampled into sampling units. There is a wide range of different methods of sampling, none of which can be described as the 'correct' or 'best' method. Stratification affects random sampling error, since only variations within each stratum will give rise to sampling error. With some material, significant gains in accuracy can be obtained if differential sampling fractions are used for the various strata. Multi-stage sampling is less accurate than a sample containing the same number of final-stage units which has been selected by a suitable single-stage method. In quota sampling, interviewers are given definite quotas of people in different social classes, or different age-groups, and are instructed to obtain the required numbers of interviews in each quota.
